Pediatric-specific issues

Designing and conducting RCTs can be challenging, and pediatric trials have additional considerations. Below are some pediatric-specific issues to keep in mind that have been identified by researchers in child health.

Blinding

• Additional groups to be blinded (i.e., parents): While in general it is important to consider how the participants and study personnel will be blinded to the assigned intervention groups, in pediatric trials it is also important to remember that there will be another group to address - parents or caregivers. When blinding parents is possible, this will reduce the risk that their knowledge of their child's intervention group will change their behaviour as related to the study outcomes.
• Coming up with a matching placebo formulation (e.g., suspensions): Often the default method of blinding includes the use of a matching placebo. If the medication can be administered as a tablet or capsule this may be straightforward, but it will be important to consider whether the child will be able to swallow a pill or whether a different formulation may be necessary. In these cases, it may be more difficult to ensure that the placebo is, in fact, 'matched.'
• Ethical approval for placebo/sham treatment: According to the Canadian Tri-Council Policy Statement 2 (TCPS2), new interventions should generally be tested against established best therapy, unless there is compelling justification for the use of a placebo. While some trials in adults have gone to great lengths to ensure blinding (e.g., sham surgical procedures), it is unlikely that these would be approved in children.
• Increased response to placebo: There is evidence that children show a greater placebo effect than adults (Rheims et al. PLoS Med 2008 [PubMed]), suggesting that expected effect sizes based on results extrapolated from adults are likely to be narrowed, resulting in false negative results (type II error). This could have implications for choice of comparator, and by extension, blinding.

Outcome measures

• Differing outcome measures by age: Due to differences in developmental and physiological stages, outcome measures may not be relevant or valid to use for all study participants. If this is the case, it is important to ensure that there are no systematic differences between intervention groups (e.g., stratify by age group so that improper comparisons are not being made - for example, comparing two different instruments, rather than interventions).

General research challenges

• Fewer patients leads to less familiarity with carrying out the procedures in the protocol: Without regular enrollment of participants into a study, it is possible that a lack of familiarity with the trial procedures could lead to more frequent protocol deviations, potentially leading to bias. It is important to have a clear protocol and monitoring procedures in place to make sure that the trial is being conducted as intended.
• Acceptance of randomization within schools and classrooms: Cluster randomized controlled trials can be relevant in pediatric studies, particularly in school and classroom settings. Recruitment in these types of trials can be difficult, owing to the vulnerability of the population and the responsibilities of the relevant gatekeepers. It is the responsibility of the researcher to adequately outline the principles of equipoise and the role of randomization in the study.

Please add to this list - either with questions or with suggestions! Feel free to leave details of your own case study, to be used as an example to help others, or to get feedback on how you could address a certain scenario. To leave a comment, use the 'Discussion Posts' icon in the grey bar on the top right-hand side of your screen.